中文摘要：

基于亮度均值减损对比归一化（MSCN）系数统计特性及其8方向邻域系数间的相关性,提出了一种通用无参考图像质量评价方法.首先,分别利用非对称广义高斯分布（AGGD）模型拟合MSCN系数及其8方向邻域系数,并估计相应AGGD模型参数作为亮度统计特征;其次,计算8方向邻域MSCN系数间的互信息（MI）,作为描述方向相关性的统计特征;进而,分别利用支持向量回归机（SVR）和支持向量分类机（SVC）构建无参考图像质量评价模型和图像失真类型识别模型;最后,在LIVE等图像质量评价数据库上进行了算法与DMOS的相关性、失真类型识别、模型鲁棒性及计算复杂性等方面的实验。实验结果表明,本文方法的评价结果与人类主观评价具有高度的一致性,在LIVE图像质量评价数据库上的斯皮尔曼等级相关系数（SROCC）和皮尔逊线性相关系数（PLCC）均在0.945以上;而且,图像失真类型识别模型的识别准确率也高达到92.95%,明显高于当今主流无参考图像质量评价方法。

英文摘要：

Based on the statistical properties of mean subtracted contrast normalized（MSCN）coefficient and the correlation between MSCN neighborhood coefficients along 8directions,ageneral-purpose noreference image quality assessment（NR-IQA）method is proposed.Firstly,asymmetric generalized gaussian distribution（AGGD）models are used to fit MSCN coefficient and its neighborhood coefficients along 8directions respectively,and the parameters of those AGGD models are estimated as the statistical characteristics of luminance.Secondly,the mutual information（MI）between neighborhood MSCN coefficients along 8directions are calculated as the statistical characteristics of directional correlation.Moreover,support vector regression（SVR）and support vector classifier（SVC）are used to construct the NRIQA model and the image distortion type recognition model,respectively.At last,in order to analyze the correlation with differential mean opinion score（DMOS）,the classification accuracy and the computational complexity,a large number of simulation experiments are carried out in the LIVE image quality evaluation database.The simulation results show that this method is suitable for many common distortions and consistent with subjective assessment,and the Spearman′s rank ordered correlation coefficient（SROCC）and the Pearson′s linear correlation coefficient（PLCC）in LIVE image quality evaluation database are more than 0.945.In addition,the recognition accuracy of the recognition model is up to 92.95%and significantly superior to all present NR-IQA methods.